TVQA+: Spatio-Temporal Grounding for Video Question Answering
TL
TL;DR
By performing this joint task, the proposed Spatio-Temporal Answerer with Grounded Evidence with STAGE model is able to produce insightful and interpretable spatio-temporal attention visualizations.
Abstract
We present the task of Spatio-Temporal Video Question Answering, which requires intelligent systems to simultaneously retrieve relevant moments and detect referenced visual concepts (people and objects) to answer natural language questions about videos. We first augment the TVQA dataset with 310.8K bounding boxes, linking depicted objects to visual concepts in questions and answers. We name this augmented version as TVQA+. We then propose Spatio-Temporal Answerer with Grounded Evidence (STAGE), a unified framework that grounds evidence in both spatial and temporal domains to answer questions about videos. Comprehensive experiments and analyses demonstrate the effectiveness of our framework and how the rich annotations in our TVQA+ dataset can contribute to the question answering task. Moreover, by performing this joint task, our model is able to produce insightful and interpretable spatio-temporal attention visualizations.
References69 items
1
LXMERT: Learning Cross-Modality Encoder Representations from Transformers
2
Gaining Extra Supervision via Multi-task learning for Multi-Modal Video Question Answering
3
Progressive Attention Memory Network for Movie Story Question Answering
4
GQA: A New Dataset for Real-World Visual Reasoning and Compositional Question Answering
5